This week's book giveaway is in the OO, Patterns, UML and Refactoring forum. We're giving away four copies of Refactoring for Software Design Smells: Managing Technical Debt and have Girish Suryanarayana, Ganesh Samarthyam & Tushar Sharma on-line! See this thread for details.
I have heard a bit about primary key generators lately on the list.I am interested in writing a simple one. What are the goals a good generator should accomplish? Any pitfalls to watch out for? Maybe someone could point out some good resources on the web? Thanks, ELahe
A primary key generator is "good" when it generates unique keys. Otherwise it's defective. The most common cause of failure is in multi-threaded systems where the process isn't atomic and the same key can be generated twice because the internal value wasn't updated before another request came in. The easiest generation technique is just to use an integer and increment it for each new row added to the table. Many DBMS's have the ability to do this for you automatically. When working with CMP EJB's, there's a problem with this, though, since CMP EJBs expect ALL values for a newly-created EJB to be supplied by the constructor and not acquired from the DBMS. TO get around this, I use a special EJB that has a "createNewKey()" method. Because it's an EJB, the atomicity of the operation is ensured by the EJB transaction mechanism.
Customer surveys are for companies who didn't pay proper attention to begin with.